划分层次限制的快速帧间预测算法

通用视频编码（Versatile Video Coding，VVC）是正在探索中的下一代视频编解码标准，在新标准的制定过程中，加入了许多新技术，在提升编码性能的同时，增加了编码复杂度。针对这种情况，通过对新标准编码过程中帧间预测单元划分算法的研究发现，在VVC进行帧间单元划分时，进行了多余的更深层次的划分，从而提高了编码复杂度。因此提出了一种划分层次限制的快速帧间预测算法，使单元划分提前结束，避免了多余的划分层次。实验结果表明，新算法在RA配置下，在增加1.58%的压缩率，损失0.036 2的图像失真度的情况下，编码复杂度降低了46.39%，从而验证了优化算法能有效降低编码复杂度。

Fast Inter-Frame Prediction Algorithm with Partition Level Limitation

Versatile Video Coding（VVC） is the next generation video codec standard, during the development of new standards, many new technologies have been added to improve encoding performance, but new technologies increase encoding complexity. In view of this situation, through the study of the inter-frame prediction unit partition algorithm in the new standard encoding process, it is found that redundant deeper partition is performed when VVC inter-frame units are partitioned, this algorithm increases coding complexity, so this paper proposes a fast inter-frame prediction algorithm with limitation on the partition level, which makes the unit partition end early, and avoids unnecessary partition levels. Experimental results show that under the RA configuration, the new algorithm increases the compression rate by 1.58% and loses the image distortion of 0.0362, but the encoding complexity is reduced by 46.39%. Therefore, the optimization algorithm can effectively reduce the encoding complexity.